This content has been marked as final. Show 2 replies
The morality of the situation can not be told by anyone who doesn't know
all about what info you are sending. In my opinion, it is wrong to send
any information at all unless you have at least warned the user that you
are doing it. If you are sending simple usage info, like the number of
times a CD has been played or something with no user-specific info, then
there shouldn't be a problem. If you are sending usernames, passwords,
social security numbers, and whatnot, then you have a real obligation to
notify the user of what is being sent, and you should encrypt the data
in some way so no-one can capture your text in transit.
Now on to the technological limitations. As soon as any internet access
is attempted, anyone with a normal outgoing firewall (pretty much all of
them apart from the useless one that comes with WinXP), will know that
it is happening. Depending on the firewall program and the technical
competency of the user, they may even be able to see exactly what text
you are sending (postNetText sends its information in plain text, so it
is easily discoverable with a sniffer). That should keep you honest,
since people will likely see what you are sending, and if you are not
telling them about it, they will find out. It is also another reason to
Again, depending on the firewall program, it may send multiple warnings,
but not usually. If you prepare your users for it, then it won't be a
surprise and they will hit Accept and then everything will work nicely.
I don't know what you mean about a 'port blocker', but I'm pretty sure
that postNetText and the other network related commands will be using
port 80 (the standard HTTP port). If port 80 is blocked, then you may
need to find another approach.
I agree and your point was exactly as I had anticipated. I'm not planning on sending any confidential info or anything weird. It's just for gathering usage stats. I have a project which will be around 200,000 discs I'd love to get a number of "how many of the 200,000" were actually inserted into the CDROM drives of people.
Currently, I'm using the webXtra to pull data off the web when a user click on a "current" content section. So far, so good and no issues but the user still needs to initiate the click. Google analytics is then incorporated into the page that is being called.
I'm just trying to avoid a "warning screen" that says "Attempting to connect to the Internet" without the user knowing... I'm not looking for a sneaky tactic, but something that's relatively solid for my needs.